Show simple item record

dc.contributor.authorNi, Yingen
dc.date.accessioned2017-12-31T07:00:23Zen
dc.date.available2017-12-31T07:00:23Zen
dc.date.issued2016-07-08en
dc.identifier.othervt_gsexam:8485en
dc.identifier.urihttp://hdl.handle.net/10919/81463en
dc.description.abstractGene regulatory networks (GRNs) provide a natural representation of relationships between regulators and target genes. Though inferring GRN is a challenging task, many methods, including unsupervised and supervised approaches, have been developed in the literature. However, most of these methods target non-context-specific GRNs. Because the regulatory relationships consistently reprogram under different tissues or biological processes, non-context-specific GRNs may not fit some specific conditions. In addition, a detailed investigation of the prediction results has remained elusive. In this study, I propose to use a machine learning approach to predict GRNs that occur in developmental stage-specific networks and to show how it improves our understanding of the GRN in seed development. I developed a Beacon GRN inference tool to predict a GRN in seed development in Arabidopsis based on a support vector machine (SVM) local model. Using the time series gene expression levels in seed development and prior known regulatory relationships, I evaluated and predicted the GRN at this specific biological process. The prediction results show that one gene may be controlled by multiple regulators. The targets that are strongly positively correlated with their regulators are mostly expressed at the beginning of seed development. The direct targets were detected when I found a match between the promoter regions of the targets and the regulator's binding sequence. Our prediction provides a novel testable hypotheses of a GRN in seed development in Arabidopsis, and the Beacon GRN inference tool provides a valuable model system for context-specific GRN inference.en
dc.format.mediumETDen
dc.publisherVirginia Techen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectNetwork inferenceen
dc.subjectsignal transduction pathwaysen
dc.subjectgene expressionen
dc.subjectsupport vector machinesen
dc.titleA Machine Learning Approach to Predict Gene Regulatory Networks in Seed Development in Arabidopsis Using Time Series Gene Expression Dataen
dc.typeThesisen
dc.contributor.departmentComputer Scienceen
dc.description.degreeMaster of Scienceen
thesis.degree.nameMaster of Scienceen
thesis.degree.levelmastersen
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen
thesis.degree.disciplineComputer Science and Applicationsen
dc.contributor.committeechairGrene, Ruthen
dc.contributor.committeechairHeath, Lenwood S.en
dc.contributor.committeememberLi, Songen


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record